Hurricane Sandy fizzled out over Quebec Wednesday morning, leaving a trail of devastation along the US East Coast and into southern Ontario. As I write this, Sandy’s death toll stands at 132 people – 71 in the Caribbean and 61 in the United States. Since making landfall in the US, it flooded the New York City subway system, left 8 million people without electricity (6.5 million of whom still lack it), destroyed the Atlantic City boardwalk, and shut down the New York Stock Exchange for 2 days. Estimates of its economic damage are up to $50 billion, making it the 2nd most expensive storm in US history, after Hurricane Katrina, which devastated New Orleans.
Figure - Hurricane Sandy seen Monday morning as it strikes the US. Image courtesy of NASA.
Sandy is a taste of things to come. As our planet warms – due to our activity – hurricane intensity will rise. Indeed, it’s rising already.
Hurricanes are giant heat engines. While over the sea – where all hurricanes are born and strengthen, and where most die out without ever touching land – they gain intensity from warm sea surface temperatures. Sandy grew so strong in large part because the ocean surface off the US East Coast this summer was 5 degrees Fahrenheit warmer than the norm. That high temperature drives more evaporation of sea water into the air above it, and more direct energy transfer into that air, adding power to the moist, destructive cyclone that is a hurricane.
Hurricane modelers have long understood this. And so it’s no surprise that hurricane models that take into account changing climate predict more total energy in future hurricanes as the planet warms. To be clear, models of the US eastern seaboard are divided on whether we should expect more hurricanes. Some predict more, some predict the same number, and some predict fewer. But a common factor is that hurricane models almost universally expect more total hurricane power - stronger winds, longer lifetimes, larger storms. That hurricane power – the total energy of hurricanes – is their total ability to destroy.
Modeling weather phenomena is incredibly challenging. We should trust models in proportion to how well they predict actual reality. To bolster our confidence in these hurricane predictions, then, we can look at how hurricane activity has changed as the planet has warmed. That historical data validates what the models say. Since the beginning of the 20th century, the number of “major hurricanes” in the North Atlantic has more than doubled.
Figure - As the planet has warmed, the number of tropical storms, hurricanes, and major hurricanes in the North Atlantic has risen. Image courtesy of GlobalWarmingArt.com
And the total destructive energy of hurricanes has risen since the 1970s, in a way that correlates extremely well with sea surface temperatures. Climate scientist Kerry Emanuel used records of North Atlantic hurricanes and North Atlantic sea surface temperatures to study how the two relate. What he found, shown in the graph below, is that the two move almost in lock step.
Figure - Total hurricane power (green line and right axis) moves almost in lock step with sea surface temperatures (blue line and left axis). Image from Emanuel (2007).
The most sobering thing about the graph above, however, is not the way the lines move in lock step. It’s the extreme sensitivity of hurricane energy to temperature. From 1970 to 2005, sea surface temperatures rose by 0.8 degrees Celsius. That change – hardly something that seems dramatic – led to roughly a quadrupling of total hurricane energy in the area. Look again at the two scales in the graph. The right scale covers a dramatically different range than the left. Small changes in temperature lead to massive changes in hurricane intensity. With sea surface temperatures projected to continue rising, particularly in coastal regions close to land, the total energy available to tropical storms is likely to soar in the decades ahead.